1. Field Of The Invention
The present invention generally relates to an apparatus and method for modulating optical energy and, more particularly, to an apparatus and method for controlling the optical energy, e.g. within a laser cavity by using and controlling dc voltages by means of light activated switches. The present invention further relates to a control system and method for controlling a laser system to enable the system to perform Q-switching, mode-locking and/or cavity dumping. The present invention also relates to an apparatus for the production of short laser radiation pulses using cavity dumping techniques. The present invention also relates to methods of Q-switching a laser system at speeds on the order of about a few hundred picoseconds.
2. Description Of The Prior Art
Optical modulators, used, for example, to control the output of lasers, are generally controlled by electrical signals. The limitation on the application of an electrical signal to a modulator in terms of speed, multiple pulse switching, volume, etc., have a direct consequence on the performance of laser systems. The present invention can be employed in lasers to generate optical pulse of large power (e.g. multi-megawatt) which have an extremely short duration. This is a regime of operation which has not previously been addressed by laser systems. The present invention provides for Q-switching a laser system at speeds on the order of about a few hundred picoseconds, a new capability requirement which has arisen with the introduction of solid state lasers having an optical length in the millimeter range. The present invention also enables mode-locking and cavity dumping, resulting in the generation of optical pulses in the picosecond regime.
Several methods have been developed for producing pulses from lasers in a controlled manner, including Q-switching, mode-locking cavity dumping, gain switching, self-injection and various combinations of these methods.
Conventional optically pumped solid state lasers (i.e. lasers based on an active medium comprised of active ions contained in a crystalline lattice, e.g. ruby, neodymium:yttrium aluminum garnet, etc.) produce pulses through one of several processes. As described by R. W. Hellwarth ("Advances in Quantum Electronics," 1961) Q-switching involves raising the cavity Q from low to high when there is a maximum population inversion in the active medium. This produces a pulse having a typical duration of a few round-trip transit times of the laser cavity. Mode-locking, where several longitudinal cavity modes are locked together in phase, creating a temporal interference such that the laser produces pulses at a rate equal to the inverse of the round-trip transit time of the laser cavity, and with a duration of some small fraction of the cavity round-trip transit time, is described by G. H. C. New, "The Generation of Ultrashort Pulses," Rep Prog Phys. 46 877 (1983). Cavity dumping, where the cavity Q is decreased from a high value, (i.e., where the circulating optical power is high) to a low value (i.e., where the circulating optical power can quickly escape from the laser cavity) is described by R. B. Chesler, et al., J. Appl. Phys. 42 1028 (1971). If the laser is not mode-locked, the pulse lengths produced in this method can be as short as the laser cavity round-trip transit time and, if the laser is mode-locked, considerably shorter.
Gain switching is described by A. Owyoung, et al., "Gain Switching of a Monolithic Single Frequency Laser-Diode-Excited Nd:YAG Laser," Optics Lett 10 484 (1985), in which the gain of the laser is quickly increased to a high level through pulsed pumping. The intracavity radiation level quickly builds up to saturate the gain in the laser and then escapes. Typical pulse lengths can be a few laser cavity round-trip transit times.
"Self-injection" or "cavity-flipping" is a known technique which involves the use of a polarization modulator for "flipping" the polarization of the intracavity radiation in a laser in order to generate a pulse whose duration is less than the cavity round trip time, as described by C. H. Brito-Cruz, et al., "The Self-Injected Nonmode-locked Picosecond Laser," IEEE Journal Quant Electron QE-19 573, (1983). This pulse is then regeneratively amplified before being dumped out. During the amplification process, pulse shortening techniques may be applied to reduce the pulse from the nanosecond regime to the picosecond regime.
A technique known as pulse transmission mode ("PTM"), where the cavity Q is kept low to build up a large gain in the active medium and is then switched high to transfer the stored energy into the optical field, is described by A. A. Vuylsteke, "Theory of Laser Regeneration Switching, J. Appl. Phys. 34 1615 (1963). At the point of maximum optical field, the cavity Q is reduced to a low value to allow the optical energy to escape quickly. This technique can be viewed as a combination of Q-switching and cavity dumping.
Microchip lasers, such as that described by A. Mooradian in U.S. Pat. No. 4,860,304, can be operated in a Q-switched mode, as described by J. J. Zayhowski and A. Mooradian in U.S. Pat. No. 5,132,977, which results in the generation of pulses shorter than 300 ps in duration. These Q-switched pulses are considerably shorter than those produced by conventional Q-switched lasers due to their extremely short cavity length, which is on the order of about 1 mm. However, the energy output from this type of laser is severely restricted due to the small volume of active material employed. The short Q-switching build up time in the microchip laser requires that the Q-switch activation be extremely fast; this can be achieved using the fast switching described in this invention. Shorter pulses than can be generated using Q-switching alone can be produced using the feedback Q-switch/cavity dumping technique described herein. This will allow the generation of pulses in the sub 100 ps range, a pulse duration range which heretofore has only been accessible using the technique of mode-locking.
Operation of arrays of microchip lasers, described by A. Mooradian in U.S. Pat. No. 5,115,445, will result in an increase of the total pulse energy to the level of a few millijoules. The present invention will enable many millijoules to be generated in the subnanosecond regime.
Conventionally, electro-optic modulators have been used in systems to control optical beams in applications such as signal modulation for optical communication and in the control of solid state lasers. When controlling solid state lasers, electro-optic modulators have been used in systems which enable Q-switching, cavity dumping, or mode-locking of the lasing system. Present electro-optic modulators, however, must be configured for each application; e.g., Q-switching, cavity dumping, or mode-locking. To perform each different application, the configuration of the modulator and its associated system must be altered.
The most common form of intracavity electro-optic modulation is Q-switching with a system that uses, for example, a Pockels cell. In a pulse reflection mode, i.e. normal Q-switching, the Pockels cell is used to change the cavity Q once during a pulse sequence. Other applications using a Pockels cell may also require the voltage applied to the Pockels cell to be turned on and off. These other applications include PTM operation, pulse slicing, optical gating, and single pulse selection. An advantage of PTM (cavity dumped) operation is that the duration of the optical pulse is limited by the cavity length, whereas the duration of a Q-switched pulse is determined by the gain characteristics and cavity decay time. Various approaches have been employed to provide turn-on and turn-off capabilities using a Pockels cell, including the use of two krytrons or a microwave tube. These approaches, however, are limited as to risetime, repetition rate and lifetime, or require complicated switched power supplies.
As disclosed in U.S. Pat. No. 3,917,943 to Auston, it is recognized that a photoconductive switch is a relatively fast, electrical switching device. The use of a photoconductive switch to control a Pockels cell to achieve a risetime of 25 ps was disclosed by LeFur, et al. in Appl. Phys. Lett. 28 21 (1976), and further detailed in Mourou and Knox, Appl. Phys. Lett 35 492 (1979) and Agoibaelli, et al., Appl. Phys. Lett 35 731 (1979).
Margulis, et al. in Optics Comm. 35 153 (1980) disclosed a photoconductive switch controlling a Pockels cell to actively mode-lock a coumarin dye laser. The signal for controlling a GaAs photoconductive switch was obtained from a second mode-locked laser. The short carrier lifetime of 100 ps allowed the Pockels cell to recover during the 5 ns round-trip transit time. Another application of photoconductive switch control discussed by M. J. P. Payne and M. W. Evans, Paper TUB16, Proceedings of Confr. on Lasers & Optics, 1984 Anaheim, Calif., was the use of negative feedback to extend the pulse length of a Q-switched Nd:YAG laser. In this case, 10% of the intracavity radiation was extracted via a beam splitter to activate a photoconductive switch, which controlled the Pockels cell voltage.
In these earlier inventions and disclosures, the essence of using a photoconducting switch was to use the fast switching speed to discharge a charged transmission line into a second, impedance matched line which contained the electro-optic modulator. This configuration is recognized as providing the fastest possible electrical risetime on the modulator.
Use of a Pockels cell as a phase retarding element in a tunable Q-switched laser is known from U.S. Pat. No. 5,001,716. Also, a Q-switched solid-state laser structure is known from U.S. Pat. No. 4,965,803. A cavity-dumped laser using feedback is known from U.S. Pat. No. 4,841,528. Finally, the use of a pulser for pulsing a Q-switch within a laser cavity is known from U.S. Pat. No. 4,752,931.